study of syntrophic anaerobic digestion of volatile fatty acids using enriched cultures at...

T. Amani et al.Int. J. Environ. Sci. Tech., 8 (1), 83-96, Winter 2011ISSN: 1735-1472© IRSEN, CEERS, IAU

*Corresponding Author Email: [email protected] Tel.: +9821 82884372; Fax: +98 21 82883381

Received 30 June 2010; revised 11 September 2010; accepted 1 November 2010; available online 1 December 2010

Study of syntrophic anaerobic digestion of volatile fatty acids using enriched cultures at mesophilic conditions

1T. Amani; 1*M. Nosrati; 1S. M. Mousavi; 2R. K. Kermanshahi

1Biotechnology Group, Department of Chemical Engineering, Faculty of Engineering, Tarbiat Modarres University, Tehran, Iran

2Department of Biology, Faculty of Basic Sciences, Alzahra University, Tehran, Iran

ABSTRACT: Volatile fatty acids are the most important intermediates in anaerobic digestion, and their degradationsare extremely complicated thermodynamically. In this research, syntrophic anaerobic digestion of volatile fatty acidsusing enriched acetogenic and methanogenic cultures in a batch reactor at mesophilic conditions was investigated.Interactive effects of key microbiological and operating variables (propionic, butyric and acetic acids, retention time andmethanogen to acetogen populations ratio) on the anaerobic degradation of volatile fatty acids were analyzed. Acetogenicand methanogenic anaerobes in the granular sludge from an up-flow anaerobic sludge blanket reactor were enriched atmesophilic conditions within a period of four weeks, separately. Enriched cultures were mixed with known proportionsand then used in the bioreactor. Experiments were carried out based on central composite design and analyzed usingresponse surface methodology. Four parameters (final concentrations of propionic, butyric and acetic acids and biogasproduction) were directly measured as response. Also, the optimum conditions for volatile fatty acid degradation werefound to be 937.5 mg/L, 3275.5 mg/L, 2319.5 mg/L, 45 h and 2.2 proportions for propionic acid, butyric acid, aceticacid, retention time and methanogen to acetogen populations ratio, respectively (corresponding to maximum volatilefatty acid removal efficiencies and biogas production). The results of the verification experiment and the predictedvalues from the fitted correlations at the optimum conditions were in close agreement at a 95% confidence interval. Thepresent study provides valuable information about the interrelations of quality and process parameters at differentvalues of microbiological and operating variables.

Keywords: Acetogens; Acetogenesis reactions; Methanogens; Methanogenesis reactions

INTRODUCTIONThe anaerobic digestion process is a sequential,

complex biochemical process, in which organiccompounds are mineralized to biogas (mainlyconsisting of CH4 and CO2) through a series of reactionsmediated by several groups of microorganisms. Thisprocess, from the view point of microbiology, followsfour major steps: hydrolysis, acidogenesis,acetogenesis and methanogenesis (Gerardi, 2003;Metcalf and Eddy, 2004; Bitton, 2005; Nwuche andUgoji, 2008; Nwuche and Ugoji, 2010). Of the severalintermediate stages, acetogenesis and methanogenesisare the most critical steps, in which acetogens(propionate and butyrate degrading bacteria) andmethanogens (hydrogenotrophs and aceticlasts) form

special constructions and are interrelated in what iscalled a “syntrophic interaction” (Schink and Stams,2005). Propionate and butyrate are the most importantintermediates in the syntrophic reactions, and becauseof thermodynamic restrictions, their degradations areregarded as the rate limiting steps in the anaerobicdigestion. They often accumulate in the anaerobicdigester when the process becomes unstable (Kida etal., 1993; de Bok et al., 2004; Wong et al., 2009; Shahet al., 2009; Liu et al., 2010; Hooshyari et al., 2009).The anaerobic oxidations of propionate and butyrateto acetate, CO2 and H2 are highly endergonic(DG°

Propionate= +76.1 and DG°Butyrate= +48.1 kJ/mol at 25

°C) and do not occur naturally and thermodynamically,but these reactions can be accomplished by thesyntrophic cooperation of propionate and butyrate-

T. Amani et al.

84

oxidizing bacteria and H2/formate-scavenging partners,which maintain a low H2 partial pressure (Schink andStams, 2005). The anaerobic conversion of butyrate isnot as unfavorable thermodynamically as theanaerobic conversion of propionate, and i tsaccumulation rarely has been reported in the literature.

Studies on the volat ile fatty acid (VFA)conversions have been mainly focused on thesyntrophic association in the co-cultivation ofacetogens with methanogens (Plugge et al., 2002; deBok et al, 2005; Kosaka et al., 2006; Babel et al., 2007;Tatara et al., 2008). In these studies, the effects of allVFAs, especially propionate, on the activity ofacetogens and methanogens have been considered.Several investigators have described the toxic effectsof VFAs in the anaerobic digestion process(Pullammanappallil et al., 2001; Han et al., 2005; Gallertand Winter, 2008; Uneo and Tatara, 2008), but theextent of this inhibition has not been properly studied.The main aim of this research was to investigate,analyze and model the syntrophic anaerobicdegradation of VFAs and the performance of theenriched acetogenic and methanogen icmicroorganisms.

It is often necessary to introduce enriched seedingmicroorganisms to start up anaerobic digesters. Theenriched cultures improve the syntrophic reactions.As another contribution in this study, a novel methodwas applied to enrich the acetogens and methanogensutilizing specific substrates. In the last few years,response surface methodology (RSM) has beenapplied to analyze, optimize and evaluate interactiveeffects of independent factors in numerous chemical,biochemical and bioenvironmental processes (Wanget al., 2005; Altaf et al., 2006; Aktaþ et al., 2006;Zinatizadeh et al., 2006; Ghorbani et al., 2008).Analysis and modeling of anaerobic digestionprocesses, particularly syn trophic anaerobicdegradation of VFAs, using RSM has not yet beenreported. In this research, the RSM was used toanalyze and model the process with respect to thesimultaneous effects of five microbiological andoperating variables (propionic acid (HPr), butyric acid(HBu), acetic acid (HAc), methanogen to acetogenpopulations ratio (M/A) and retention time (RT)) andfour parameters (HPr, HBu, HAc and biogasproduction) were assessed as responses. Thesignificant factors and a continuous response surfaceof the main parameters were developed to provide an

optimal region that sat isfies the processspecifications. This research was carried out in Bio-pilot Laboratory, Biotechnology Group, ChemicalEngineering Department, Tarbiat Modares Universityduring November 2009 to February 2010.

MATERIALS AND METHODSEnrichment

To enrich the acetogens and methanogens, thegranular sludge from an up-flow anaerobic sludgeblanket (UASB) reactor that is employed in thetreatment of dairy wastewater was utilized. The pH ofthe granular sludge was 7.4, and the concentrations ofvolatile suspended solid (VSS) and total suspendedsolid (TSS) were 67.2 g/L and 92.4 g/L, respectively. Toenrich the anaerobes, the minimal medium containedthe following: KH2PO4 (0.6 g/L), Na2HPO4 (1.33 g/L),NH4Cl (0.3 g/L), CaCl2.H2O (0.11 g/L), MgCl2.6H2O (0.1g/L), NaHCO3 (4.0 g/L), Na2S.9H2O (0.025 g/L), FeCl2(0.45 g/L), H3BO4 (0.04 g/L), ZnCl2 (0.032 g/L), CuCl2(0.0063 g/L), MnCl2 (0.0295 g/L), CoCl2 (0.0305 g/L),NiCl2 (0.0062 g/L), cysteine hydrochloride (0.05 g/L)and the yeast extract (0.05 g/L). The medium cultureswere prepared with distilled water. Before inoculation,O2 was removed from all the culture vessels and mediaby sparging with the N2 gas for 10 min. The enrichmentof propionate- and butyrate-degrading bacteria andmethanogens was performed at 37±1 °C. They wereenriched in three Erlenmeyer flasks with workingvolumes of 1 liter, independently. The carbon sourcesfor enrichment of propionate-, butyrate-degradingbacteria and aceticlasts were HPr (3000 mg/L), HBu(4000 mg/L) and HAc (5000 mg/L), respectively.

To enrich the anaerobic hydrogenotrophs,hydrogen was produced by the electrolysis of tap water.The schematic representation of the hydrogengenerator is shown in Fig. 1. This system consists of arectifier (maximum voltage: 35 V, and maximum current:15 A), 2 stainless-steel electrodes (dimensions:2×300×30 mm3) and a vessel containing 8 liters of tapwater. Conductivity of water was increased by theaddition of 3 to 4 milliliters of concentrated sulfuricacid (98 %). Under these conditions the maximumelectrical current passing through the water was 3 A.H2 gas was produced and collected at the negativeelectrode, followed by venting it to the vesselcontaining the granular sludge and HAc as the carbonsource. Enrichment of hydrogenotrophs with the H2gas was performed at a flow rate of 95 mL/h. The initial

T. Amani et al.

85

Int. J. Environ. Sci. Tech., 8 (1), 83-96, Winter 2011

pH was adjusted at 7±0.1 with NaOH. The temperatureconditions were controlled by using a water bath. Themagnetic stirrer was used to mix contents of the vesselsduring the enrichment process. Mixing was performeddaily with the intensity of 50 rpm and duration of 5 min.The initial enrichment cultures were started by adding200 mL of the granular sludge to 800 mL of the minimalmedium with the carbon source (20 % (v/v) inoculation).Every week, samples of each vessel were drawn, andtheir pH and VFA concentrations were assayed. Theinoculated culture (200 mL) was then transferred to thefresh medium (800 mL). The removal efficiencies of VFAs(more than 95 % removal efficiency) were considered asthe criteria to perform the enrichment of the anaerobicmicroorganisms. After a few transfers, all done accordingto the defined criteria, the enrichment processes wereconsidered complete and the anaerobic microorganismscontinued to remain in the cultures.

SubstratesThe synthetic wastewater, containing pure HPr, HBu

and HAc (Merck Inc., Germany), as well as tap waterwith different concentrations (low to high loads), wereused as the major carbon sources and electron donorsin this study. To supply adequate nitrogen andphosphorus for the microorganisms, chemical oxygendemand (COD):N:P ratio was maintained at 100:4:1. O2was removed by N2 sparging for 10 min before feedingto the bioreactor. The pH of the feed and within thereactor was not regulated throughout the experiment(4.5–5.5). To increase the alkalinity, NaHCO3 (4 g/L) wasadded to the feed.

Fig. 1: Schematic representation of hydrogen generator set-up: (1) rectifier, (2) positive electrode, (3) negative electrode, (4)electrolysis vessel and (5) enrichment culture vessel.

Batch reactor and operating conditionsA flow diagram of the experimental set-up is shown

in Fig. 2. The batch reactor was a glass cylinder witha diameter of 10 cm and a height of 15 cm (workingvolume 1 L). Effluent samples were drawn from thebottom of the reactor using a sampling port. Prior tothe experiments, a 200 mL (20 % (v/v) inoculation)mixture of the enriched cultures (methanogens andacetogens) with defined M/A was used to seed thebatch reactor. For starting up the bioreactor, the ratioof methanogens to acetogens (M/A) was taken asthe relative amount of their VSS concentrations. Anelectrical heating tape (heating capacity: 40 W/m) wasattached to the outside surface of the reactor and atemperature probe was connected to the transmitter.The temperature of the reactor was set to themesophilic (37±1 °C) condition. To ensure efficienttransfer of the intermediates and to release gasbubbles trapped in the medium mixing was performedwith an intensity of 50 rpm, duration of 5 min per each10 h, using a magnetic stirrer. The produced biogaswas vented out the top of the bioreactor through aconnecting pipe and was collected by the water-displacement method. Generally, concentrations ofHPr, HBu and HAc, RT, temperature, mixing, pH andthe relative population of syntrophic microorganisms,all affect the anaerobic digestion of the VFAs. Toinvestigate the syntrophic anaerobic digestion ofVFAs, concentrations of HPr, HBu and HAc, RT andM/A were selected as the primary factors affectingthe study, and central composite design (CCD) wasused to design the experiments. The RSM used in thepresent study was CCD, which is the most popular

220

ON/OFF

3A Fuse

2 3

41 5

O2

H2

T. Amani et al.

86

Syntrophic anaerobic digestion of volatile fatty acids

response surface method (Anderson and Whitcomb,2000; Myers and Montgomery, 2002; Söteman et al., 2005).The levels of the factors are shown in Table 1. Each factorwas varied at five levels, whereas the other parameterswere kept constant. Consequently, 47 experiments wereconducted; 32 of them organized in a full factorial design,10 experiments were related to axial points and 5experiments were done in center points. The remainingfive involved repetition of the central idea to get a goodestimate of the experimental error. Experiments weredesigned by Design Expert (DX) Software (State-Ease Inc.,version 7.0.0), which is a Windows-compatible software.It provides efficient design of the experiments (DOEs) foridentification of the vital factors that affect the processand uses RSM to determine the optimal conditions.

Factor Low Axial Low Factorial Center High Factorial High Axial (– ) (–1) (0) (+1) (+ ) A: RT (h) 11 25 35 2.5 59 B: M/A ratio (g VSS methanogens/g VSS acetogens)

1.1 1.7 2.1 2150.0 3.1

C: HPr concentration (mg/L) 101 937 1543.5 4980.0 2968.0 D: HBu concentration (mg/L) 0 2031.9 3506 5685.0 70011.9 E: HAc concentration (mg/L) 0 2319.5 4002.3 8004.5

Table 1: The level of factors in CCD

Analytical methodsMethane concentration in produced biogas was

determined with Figaro TGS 2611 methane sensor (madein USA). The sensing element is comprised of a metaloxide semi-conductor layer formed on an aluminasubstrate of a sensing chip together with an integratedheater. In the presence of a detectable methane, thesensor’s conductivity increases depending on themethane concentration. A simple electrical circuit canconvert the change in conductivity to an output signalwhich corresponds to the methane concentration.

The liquid samples were first centrifuged at 10,000rpm for 15 min, followed by the HCl acidification, andfinally assayed. The VFAs (HAc, HPr and HBu) werequantified using the Agilent gas chromatograph (model

220V

6

3

2

220V 1

Batch Bioreactor 4

5

7

ON/OFF

220V

Fig. 2: Schematic flow diagram of experimental set-up: (1) magnetic stirrer, (2) electrical heating tape, (3) temperature probe,(4) biogas collector vessel, (5) methane sensor, (6) temperature controller and (7) methane sensor transmitter.

T. Amani et al.

87


Time (Week)

VFA (mg/L) VFA removal

COD (mg/L) COD Removal

Final pH

Initial Final (%) Initial Final (%) Pr.-degrading bacteria (carbon source HPr)

1st Week 2nd Week 3rd Week 4th Week

3000.0 3000.0 3000.0 3000.0

1882.3 576.5 110.1 106.3

37 81 96 96

4451.2 4451.2 4451.2 4451.2

3069.9 1104.6 275.7 265.3

30 75 94 94

7.4 7.9 7.9 7.9

Bu.-degrading bacteria (carbon source HBu)


4000.0 4000.0 4000.0 4000.0

2478.7 525.3 75.2 74.2

38 87 98 98

7268.0 7268.0 7268.0 7268.0

4846.7 1047.8 303.6 297.9

33 86 96 96

7.7 7.9 8.0 8.0

Methanogens (carbon source HAc)


5000.0 5000.0 5000.0 5000.0

2293.4 785.6 619.8 196.2

54 84 88 96

5286.6 5286.6 5286.6 5286.6

2394.8 809.7 750.7 285.4

55 85 86 95

7.4 7.6 7.6 7.7

Table 2: The results of enrichment process.

7890), equipped with an auto-injector (7683 B series), aflame ionization detector (FID) (H2 flow rate: 35 mL/min, air flow rate: 350 mL/min) and a Chrompack Cp-Wax 52 CB fused-silica column (25 m × 0.32 mm i.d. and0.2 ìm film thickness). The injector and detectortemperatures were kept constant at 240 °C and 280°C,respectively. Helium (He) was used as the carrier gasat a flow rate of 3 mL/min and makeup flow rate of 5 mL/min. The oven temperature was programmed at 40 °Cfor 4 min, raised to 180 °C at 30 °C/min, and then held at180 °C for 1 min. The VSS, TSS and COD weredetermined according to the standard methods (APHA,1992). The pH was measured using the Metrohm 620pH meter (made in Germany).

RESULTS AND DISCCUTIONEnrichments

Concentrations of COD, VFAs (HPr, HBu and HAc)and pH, during the enrichment processes weremeasured at the end of each transfer. The initial andultimate concentrations of COD, VFAs and pH in theenrichment cultures of propionate- and butyrate-degrading bacteria (acetogens) and methanogens areshown in Table 2. The results of COD and VFA removalefficiencies were in good agreement with each other.Increases of HPr, HBu and HAc removal efficiencies ordecreases of their ultimate concentrations werestepwise gradual. These results showed that after fourtransfers (four weeks) VFA removals were greater than95 %. Thus, according to the defined criteria, it couldbe concluded that acetogens and methanogens in thegranular sludge prevailed in the cultures.

During the enrichment process the initial granuleswere broken down gradually, and in the end of the fourthweek, almost all of the microorganisms were completely

dispersed. The reason is that only one VFA was usedfor the enrichment of each group of microorganisms andthe syntrophic reactions were not complete. Therefore,the enriched cultures obtained by the reaction, especiallyacetogens, were not granular and were fine or dispersedat the end of the enrichment processes. Enriched cultures(with known ratio) were thickened and then used for thesyntrophic anaerobic digestion of VFAs in the bioreactor.The concentrations of the propionate-degrading bacteriain the enriched cultures were VSS=68.4 g/L and TSS=78.2g/L. For butyrate-degrading bacteria, the concentrationswere VSS=71.2 g/L and TSS=81.3 g/L; and formethanogens, they were VSS=74.3 g/L and TSS=84.5 g/L. To verify the performance of each enriched cultureagainst a different feed, the enriched cultures wereinoculated with other carbon sources, i.e., HPr was addedto the enriched sludge of butyrate-degrading bacteriaand to the enriched sludge of methanogens. Moreover,HBu was added to the enriched sludge of propionate-degrading bacteria. The minimal medium, temperatureand pH were the same as during the enrichmentprocesses, but the incubation time was 10 days. Theconcentrations of the initial and ultimate COD are shownin Table 3. The results showed that just before 10 daysin culture, low COD removal efficiencies of 19, 11 and 21percent were observed for the propionate-, butyrate-degrading bacteria and methanogens, respectively.These results again revealed that the enriched culturespredominantly contained the propionate- and butyrate-degrading bacteria, as well as the methanogens.

Statistical analysisForty seven experiments were designed using CCD.

The experimental conditions and their responses forthe mesophilic anaerobic digestion process are shown

T. Amani et al.

88

Table 3: The results of enrichment processes validity

Enriched Culture Carbon Source COD (mg/L) COD Removal Efficiency Initial Final (%) Pr.-degrading bacteria HBu 7268.0 5886.1 19 Bu.-degrading bacteria HPr 4451.2 3961.6 11 Methanogens HPr 4451.2 3515.4 21

Run Factors Responses RT (h) M/A

(mg/L) HPr

(mg/L) HBu

(mg/L) HAc

(mg/L) HPr

(mg/L) HBu

(mg/L) HAc

(mg/L) Biogas (m/L)

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

45 45

45 45 25 25 35 45 35 25 25 35 35 35 45 45 35 25 25 25 45 35 45 35 35 45 25 25 35 45 25 25 45 59 45 35 45 45 25 35 25 11 25 25 45 35

2.5 2.5 2.5 1.7 1.7 1.7 2.1 2.5 1.1 1.7 2.5 2.1 2.1 2.1 2.5 1.7 2.1 2.5 1.7 1.7 1.7 3.1 1.7 2.1 2.1 1.7 1.7 2.5 2.1 1.7 2.5 2.5 2.5 2.1 2.5 2.1 2.5 1.7 1.7 2.1 2.5 2.1 2.5 1.7 1.7 2.1

2150.0 937.0

2150.0 2150.0 937.0

2150.0 1543.5 937.0

1543.5 937.0

2150.0 1543.5 101.0

1543.5 2150.0 937.0

1543.5 937.0

2150.0 2150.0 2150.0 1543.5 937.0

1543.5 1543.5 2150.0 2150.0 2150.0 1543.5 2150.0 937.0

2150.0 937.0

1543.5 2150.0 1543.5 937.0 937.0 937.0

2986.0 937.0

1543.5 2150.0 937.0 937.0

1543.5

4980.0 2031.9 4980.0 2031.9 2031.9 4980.0 3506.0 2031.9 3506.0 4980.0 2031.9 3506.0 3506.0 7011.9 2031.9 4980.0 3506.0 4980.0 4980.0 2031.9 4980.0 3506.0 4980.0

0.0 3506.0 4980.0 2031.9 4980.0 3506.0 2031.9 2031.9 2031.9 4980.0 3506.0 2031.9 3506.0 4980.0 2031.9 2031.9 3506.0 4980.0 3506.0 4980.0 4980.0 2031.9 3506.0

2319.5 2319.5 5685.0 2319.5 5685.0 5685.0 4002.3 5685.0 4002.3 5685.0 5685.0

0.0 4002.3 4002.3 5685.0 2319.5 4002.3 5685.0 2319.5 5685.0 5685.0 4002.3 5685.0 4002.3 8004.5 2319.5 2319.5 2319.5 4002.3 5685.0 5685.0 2319.5 2319.5 4002.3 2319.5 4002.3 5685.0 5685.0 2319.5 4002.3 2319.5 4002.3 5685.0 2319.5 2319.5 4002.3

1557.8 380.9

1708.9 1364.1 680.7

1807.3 912.2 576.2

1394.1 765.4

1405.1 658.8 62.4

1339.3 1278.2 617.1 923.4 746.9

1741.7 1436.5 1738.3 1296.3 680.8 456.7

1049.4 1587.9 1498.1 1715.5 895.6

1284.4 638.3

1190.1 607.4 821.6

1092.3 906.3 667.2 593.8 471.6

2286.7 651.9

1084.6 1891.8 774.4 424.1 901.1

3318.4 792.9

3761.3 1163.1 1996.8 4302.7 1415.4 1879.7 2864.9 4465.1 1448.5 1598.7 2291.4 6181.4 1331.3 2577.6 1522.3 4024.9 3779.2 1580.1 3881.9 2845.1 3696.1

0.0 1815.4 3401.2 1304.1 3646.7 1378.3 1377.6 2112.1 1288.0 2504.5 1278.7 1033.6 1389.1 3622.5 1906.7 1114.1 2434.3 2844.3 3085.6 4105.0 3098.1 914.3

1446.6

1740.6 871.4

3487.2 1558.2 3807.7 4865.1 2187.8 2158.4 3213.6 4343.9 3466.7 386.6

2070.4 3921.8 2589.7 1378.4 2119.5 3382.5 2288.4 3824.4 4698.1 2878.6 3820.8 2176.8 7168.9 2148.5 1757.6 2232.3 2208.7 3498.4 2598.2 1166.5 1077.8 1857.4 1045.9 2232.8 2764.1 2960.1 1435.3 3682.5 1234.9 3063.4 3961.4 1447.8 1285.0 2123.6

1486 1456 1610 1183 850 893

2060 1476 794 875

1435 1375 1375 645

1756 1920 2055 1458 995

1286 1305 904

1535 1150 1545 1276 945

1045 2145 1565 1190 1295 1985 2435 1544 2115 1912 1193 1085 1125 1713 860

1230 1513 1230 2175

Table 4: The experimental plan of VFA digestion and obtained results

T. Amani et al.

in Table 4. Because four responses were investigatedin this study, the results were inserted into DX-7software and were fitted to quadratic correlations andthen, correlations were found to be adequate for theprediction of the response variables. The analysis ofvariance (ANOVA) results for the responses have beensummarized in Table 5. The relatively high values of R2

indicate that the quadratic equations for output HPr,HBu, HAc and biogas production are well capable ofrepresenting the system under the given experimentaldomain. As can be observed in Table 5, there is verylittle difference between R2 and adjusted-R2, whichshows there is a good chance that important terms areincluded and considered in the correlations.

The correlation terms in the equations are thoseessential terms that remained after elimination ofinsignificant variables and their interactions. As seenin the table, the fitted correlations are significant witha 95 % confidence interval (p-value < 0.0001) for thefour responses. The correlation adequacy was testedthrough lack-of-fit F-tests (Montgomery, 1991). Thelack-of-fit F-statistic was not statistically significantbecause the p-values were greater than 0.05. Adequateprecision is a measure of the range of predictedresponse relative to its associated error or, in otherwords, a signal-to-noise ratio. Its desired value is 4 ormore (Mason et al., 2003). The values were found to beacceptable for the three correlations; see the adequateprecision column in Table 5 with values that are muchmore than 4. Simultaneously, low values of thecoefficient of variation (CV) for the responses indicatedgood accuracy and dependability of the experiments.

Effects of M/A and RT on VFA removalAnaerobic digestion of HPr and HBu are highly

endergonic and do not occur naturally (from the viewpoint of thermodynamic principles) in anaerobicdigesters (Schink and Stams, 2005). Small populationsof methanogens are not able to metabolize the hydrogenand HAc produced by the acetogens. Increasing themethanogenic population could be used as a methodto promote these reactions efficiently (Schmidt andAhring, 1995).

In this study, to investigate the effect of themethanogenic population on the anaerobic conversionof VFAs, the ratio was adjusted within a range of 1 to 3.Fig. 3 shows the effect of M/A on VFA removals. WhenM/A was increased from 1.1 to 2.1 (when HPr = 1543.5mg/L, HBu = 2000.8 mg/L, HAc = 2499.9 mg/L and RT =

35 h), the removal efficiencies of HPr, HBu and HAcincreased from 10, 18 and 20 to 41, 59 and 46 %,respectively. However, when M/A = 3.1, the removalrates decreased. This showed that in very high M/Aranges, the acetogens (propionate- and butyrate-oxidizing bacteria) were not sufficient to degrade highconcentrations of HPr and HBu. In this condition, highlevels of HPr and HBu inhibited the acetogenicreactions and suppressed the growth of acetogens;consequently, their degradations slowed down. Asseen in Fig. 3, higher M/A also caused HAc removal todecrease; this could be attributed to inhibition of thegrowth of the methanogenic population in the presenceof high levels of HAc and reduced pH. Accordingly, inthe high M/A ranges the activities of all anaerobesdropped off significantly.

The RT had a positive effect on VFA removal: withincreasing RT, anaerobic microorganisms adapted tolow pH and removal efficiencies improved. Fig. 4illustrates the effects of RT and M/A on anaerobicconversion of HPr. From the fitted correlations in Table5, it can be concluded that the coefficient of RT is moresignificant than the M/A coefficient for the finalconcentrations of HPr and HBu. The interaction of RTand M/A is not very significant because the coefficientof the interaction term of the correlations is small.

Effects of VFAs on HPr conversionAnaerobic oxidation of HPr is inhibited by VFAs;

the extent of this inhibition is dependent on the VFAconcentrations and the pH (Siegert and Banks, 2005).

Fig. 3: Effects of M/A on anaerobic conversion of VFAs atconstant RT


5

15

25

35

45

55

65

0.5 1 1.5 2 2.5 3 3.5

M/A

HBu HAc HPr

M/A

HBu HAc HPr

VFA

Rem

oval

( %

)

65

55

45

35

25

15

50.5 1 1.5 2 2.5 3 3.5

89

T. Amani et al.

Table 5: ANOVA results for the equations of the DX-7 for studied responsesResponse Correlations with significant terms P-value R2 Adj. R2 SD Adequate

Precision CV Press

HPr 914.64–53.22A–26.93B+459.16C+152.87D +63.03E–14.65AC+18.84BD+24.53BE+58.71CD +11.46A2+80.77B2+50.60C

<0.0001 0.9912 0.9844 61.55 53.08 5.91 3708000

HBu

1461.76–211.00A–46.11B+60.44C+1144.99D +284.73E+18.80AB+14.69AC–67.60AD–24.95BD +139.01CD – 190.91CE+48.80DE+ 148.49A2 +267.43B2+180.43C2+ 309.10D2+64.50E2

<0.0001 0.9722 0.9509 280.32 29.07 11.90 9382000

HAc

2083.76–203.48A–252.16B+285.86C+318.23D +1119.14E–34.37AB +11.05AC+20.74AD–81.28AE –146.74BE–106.10CD +17.59CE +100.18DE +109.08B2+79.09C2 +109.65D2+238.42E2

<0.0001 0.9437 0.9004 405.38 19.65 15.69 19150000

Biogas

2117.55+209.52A+103.86B–56.14C+17.68D +40.16AD+29.41AE+27.91BE–134.53CD +94.34CE–57.28DE–78.02A2 –219.17B2–219.17B2 –148.28C2–210.60D2–111.16E2

<0.0001 0.9116 0.8435 166.59 13.86 11.81 3189000

Methanogens are extremely sensitive to pH, withoptima between 6.8 and 7.4 (Turovskiy and Mathai,2006), but acetogenic microorganisms are somewhatless sensitive and can function in a wider range of pH.Generally, the presence of VFAs leads to a drop in pHin a digester, and their toxicities are steeper when thepH is below 7 (Hwang et al., 2004). Because acetogenic

Fig. 4: Effects of RT and M/A on anaerobic conversion of HPr in contour plot

degradation of HPr and HBu are extremely endergonic,their anaerobic degradations are repressedthermodynamically. Also, degradation of HPr and HBuis inhibited via a pH drop in the digester. On the otherhand, methanogenic conversion of HAc is exergonic;thus it could be expected that its inhibitory effect isdue to the pH drop, which reduces the growth of


980.468

923.235 907.542 893.612 884.26

878.575

944.005

961.781

980.468

1016.93

1053.39

5

Design Points2286.7

62.4

25.00 30.00 35.00 40.00 45.00

2.50

2.29

2.08

1.86

1.65

M/A

HPr (mg/L)

RT (h)

90

T. Amani et al.

methanogens. In fact, methanogenic oxidation of HAcstimulates HPr degradation thermodynamically. Fig. 5shows the effects of HBu and HAc on anaerobicoxidation of HPr. As seen, when concentrations of HBuand HAc were increased, the HPr removal decreased.The maximum conversion of HPr was obtained whenHBu and HAc were diminished. Data from theexperiments in Table 4 confirmed these results.Furthermore, because the mechanisms of HBu and HAcinhibition are rather different, their observed inhibitoryeffects on HPr removal were dissimilar. The coefficientof the HBu term (153) in the fitted correlation in Table 5was, as expected, greater than that of HAc term (63).Since HAc accumulated when there were high levelsof M/A, the relative inhibitory effects of HBu and HAcwere not strongly dependent on M/A; therefore, itseemed that the level of HAc inhibition with higher M/A was apparently only a little less than HBu. HPr hadthe largest inhibitory effect on the HPr removal; on theother hand, its coefficient in the fitted correlation (459)is much higher than that for HBu and HAc.

Effects of VFAs on HBu conversionAlthough anaerobic oxidation of HBu is not as

thermodynamically unfavorable as conversion of HPr,the inhibition pattern of VFAs on anaerobic conversion

of HBu is similar to HPr inhibition. Both HPr and HAcdecrease the pH of the reactor; thus, they can inhibit thegrowth of acetogenic and methanogenicmicroorganisms. Thermodynamically HPr hinders,whereas HAc enhances, the anaerobic conversion ofHBu. Accordingly, it could be expected that the inhibitoryeffect of HPr on HBu degradation is much higher thanthe inhibitory effect of HAc. The effects of HPr andHAc on HBu oxidation is shown in Fig. 6.

The results revealed that, in spite of what it would beunderstood apparently from the above, the inhibitoryeffect of HAc was greater than that of HPr. Thecoefficients of their individual terms (60, 285 for HPr andHAc, respectively) in the fitted correlation confirmedthis observation. The main reason could be the effect ofHAc on pH reduction in the digester, which seriouslyhindered the growth of anaerobes. Another reason couldbe attributed to the spatial position or juxtaposition ofthe syntrophic acetogenic and methanogenicmicroorganisms (suspended growth). In the syntrophiccultures, diffusion of dissolved intermediates (H2/formate and HAc) from the acetogens (as producers) tothe methanogens (as consumers) is the main mechanism;therefore, close microbial proximity plays an importantrole in this phenomenon. The relative spatial locationand mobility of different syntrophic organisms in the


Design Points2286.7

62.4

716.761

732.815

755.177

774.37

791.504807.855

838.899823.1

852.608 923.235907.542

893.612

878.575

5 944.005 980.468

961.781

1016.93999.382

1036.79

1053.39

1071.32

1085.4

5685.00

4843.63

4002.25

3160.88

2319.502031.90 2768.92 3505.95 4242.97 4980.00

Fig. 5: Effects of HBu and HAc on HPr removal in contour plot

HPr (mg/L)

HBu (mg/L)

HA

c (m

g/L)

91

T. Amani et al.

Fig. 6: Effects of HAc and HPr on HBu removal in contour plot

Fig. 7: Effects of RT and M/A on biogas production in contour plot

1420.05

1476.62

1523.91

1576.65

1637.94

1960.57

1799.25

1741.49

1689.27

5

5685.00

4843.63

4002.25

3160.88

2319.50

1191.54

1245.98

1315.3

1366.5

1799.25

2031.90 2768.92 3505.95 4242.97 4980.00

25.00 30.00 35.00 40.00 45.00

2.50

2.29

2.08

1.86

1.65

2249.64

2080.50

2133.2 2173.30 2206.7 2232.11

2006.76

1944.831880.25

1753.73

1627.21

5

HBu (mg/L)

HPr (mg/L)

HA

c (m

g/L)

Biogas Production (mL)

M/A

RT (h)

Design Points2435

645

Design Points6181.4

0

92

T. Amani et al.

Fig. 8: Effects of HPr and HBu on biogas production in contour plot

bioreactor may play a significant role in controlling theinhibitory effect of HPr and HAc on HBu degradation.Consequently, the movement of butyrate-degradingbacteria probably was inadequate for trouble-freediffusion of intermediates. Similar to HPr, HBu itselfhad the largest inhibitory effect on HBu removal,because its coefficient in the fitted correlation (1145) ismuch higher than that for HPr or HAc.

Biogas production from VFAsIn this study, all variables directly or indirectly

affected biogas production in the syntrophic anaerobicdigestion of VFAs, and the extents of the influenceswere dependent on their levels and interrelatedinteractions. Fig. 7 shows the effects of RT and M/Aon the biogas production. Increasing M/A enhancedthe methanogenic populations; therefore, the rate ofhydrogen consumption was increased, which led to anincrease in the biogas production. With higher levelsof M/A, the rate of hydrogen production fromacetogenesis reactions (because of the reduction ofacetogens) was decreased, and consequently biogasproduction was reduced critically. Thus, there was anoptimum M/A for the biogas production, as seen inFig. 7. The RT had a positive effect on the biogasproduction; RT and biogas production increased

simultaneously. However, in very high RT ranges, therate of biogas production was decreased and seemedto even stop. The fitted correlation in Table 5 for thebiogas production shows that the effect of RT onbiogas production was much greater than the effect ofM/A, but RT/biogas production interaction was notan important term in the correlation.

Fig. 8 illustrates the effects of HPr and HBu on thebiogas production. HPr and HBu to some extentenhanced the biogas production, but in their higherlevels, they strongly inhibited the growth of anaerobes,which led to reduction in biogas production. As a result,there were optimum amounts of HPr and HBu for thebiogas production. The presence of VFAs in theanaerobic digester leads to a reduction in pH and aninhibition of growth in anaerobes, especiallymethanogens (Turovskiy and Mathai, 2006). Themethanogenesis at low pH is attributed to the activityof hydrogen-utilizing methanogens that are likely moretolerant to the acidic conditions than the othermethanogens (Kim et al., 2004). Because methanogenicactivity is decreased in the low levels of pH, the methaneyield is decreased significantly. Fig. 9 shows methanecontent versus final pH in the anaerobic digestion ofVFAs. Although the pH in the digester was increasingduring anaerobic digestion of VFAs, because of the


937.00 1240.25 1543.50 1846.75 2150.00

2124.98

1880.25

1944.83

2052.722006.76

1753.73

1880.25

5

5685.00

4843.63

4002.25

3160.88

2319.50

1753.731880.25

2080.56

2113.91

Design Points2435

645

HPr (mg/L)

HB

u (m

g/L)

Biogas Production (mL)

93

T. Amani et al.

Fig. 9: Final pH vs. methane content for the anaerobic digestion of VFAs.

Responses Target Correlation Predicted Confirmation Experiment Confidence Interval (95%) Low High HPr (mg/L) Minimize 383.7 436.8 323.2 444.2 HBu (mg/L) Minimize 975.3 1196.4 699.8 1250.7 HAc (mg/L) Minimize 843.5 863.2 445.2 1241.9 Biogas (mL) Minimize 2077 1954 1913.8 2241

Table 6: Verification experiment at the optimum conditions

high activity of enriched cultures, the methane contentof the biogas was still low.

Maximum VFA removal and biogas productionAccording to the main objectives of this experiment

(maximum VFA removals and biogas production), theoptimum conditions were obtained to be HPr = 937.1mg/L, HBu = 3275.5 mg/L, HAc = 2319.5 mg/L, RT = 45h and M/A = 2.2. To check the accuracy of the fittedcorrelations of the optimum conditions at the 95 %confidence interval, the batch bioreactor was operatedto compare actual responses with predicted responses.Table 6 presents the results of experiments conductedat optimum conditions. The accuracy of the optimumconditions from DOE experiments was checked thatthe experimental finding was in close agreement withcorrelation predictions.

CONCLUSIONSyntrophic anaerobic digestion of VFAs using

enriched acetogenic and methanogenic cultures in a

batch bioreactor at mesophilic conditions wasinvestigated in this study. Several optimum conditionscan be considered, but the main aim of this researchwas to maximize VFA removal that their anaerobicdigestion at high concentrations thermodynamicallyis intricate or even impossible. The concentrations ofHPr, HBu, HAc, M/A and RT were selected as thecontrol factors, and HPr, HBu, HAc and biogasproduction were the main responses. Analysis of theexperimental results using RSM revealed theinterrelated interactions of the parameters. RT and M/A had positive effects on VFA removal; however,performance of the process at very high M/Adrastically decreased. In the high M/A regions, thenumber of acetogenic bacteria was not sufficient tooxidize HPr and HBu effectively. The influence of RTon VFA removal and biogas production was moresignificant than M/A. HPr and HBu inhibitedsyntrophic oxidation of VFAs thermodynamically aswell as via a pH drop. The effect of HBu on VFA removalwas more significant, which might be due to spatial


10

15

20

25

30

35

40

45

50

55

0 5 10 15 20 25 30 35 40 45 50Run

Met

hane

Con

tent

%

4

5

6

7

8

9

10

11

Fina

l pH

Methane Content pH

RunMethane pH

55

50

45

40

35

30

25

20

15

10

11

10

9

8

7

6

5

40 5 10 15 20 25 30 35 40 45 50

Fina

l pH

Met

hane

con

tent

(%

)

94

T. Amani et al.

proximity or relative positions of suspended syntrophicorganisms in the batch bioreactor. According to thefitted correlations, HPr was the most importantparameter because it influenced all the responses andreduced the performance of the process. HAcstimulated VFA removal thermodynamically, but itsinhibitory effect due to the pH drop was significant.Although syntrophic microorganisms, especiallymethanogens, are very sensitive to low pH,acetogenesis and methanogenesis occurred efficientlyin this research work, and the enriched cultures didnot lose their activity even at very low pH. Also, pHwas increased during all of the experiments in theanaerobic digestion process. The maximum VFAremovals and biogas production occurred whenHPr=937.1 mg/L, HBu=3275.5 mg/L, HAc=2319.5 mg/L,RT = 45 h and M/A = 2.2.

Syntrophic studies may be more accurate when ratiosof M/A become more accurate. This may be achievedby obtaining optimum ratios of propionate-degradingbacteria to butyrate-degrading bacteria in theacetogenic microorganisms bulk.

REFERENCESAktaþ, N.; Boyacý, Ý.H.; Mutlu, M.; Tanyolaç, A., (2006).

Optimization of lactose utilization in deproteinated wheyby Kluyveromyces marxianus using response surfacemethodology (RSM). Bioresource. Tech., 97, 2252–2259(8 pages).

Altaf, M.D.; Naveena, B. J.; Venkatshvar, M.; Kumar, E.V.;Reddy, G., (2006). Single step fermentation of starch to1(+) lactic acid by lactobacillus amylophilus CV6 in SSFusing inexpensive nitrogen sources to replace peptone andyeast extract-optimization by (RSM). Proc. Biochem., 41,465–472 (8 pages).

Anderson, M. J.; Whitcomb, P. J., (2000). DOE simplified:Practical tools for effective experimentation. Productivity,Inc. Portland, Oregon.

APHA., (1992). Standard methods for the examination ofwater and wastewater, 18th Ed. Washington, DC: Am. PublicHealth Assoc.

Babel, S.; Opiso E. M.; Removal of Cr from syntheticwastewater by sorption into volcanic ash soil. Int. J. Environ.Sci. Tech., 4 (1), 99-108 (10 pages).

Bitton, G., (2005). Wastewater microbiology, A John Wiley &Sons Inc. Publication, 3rd Ed., Hoboken, New Jersey Chap.,13, 345–369 (25 pages).

de Bok, F. A. M.; Plugge, C.; Stams, A. J. M., (2004). Interspecieselectron transfer in methanogenic propionate degradingconsortia. Water Res., 38, 1368–1375 (8 pages).

de Bok, F. A. M.; Harmsen, J. M.; Plugge, C. M.; de Vries, M.C.; Akkermans, A. D. L.; de Vos, W.M.; Stams, A. J. M.,(2005). The first true obligatory syntrophic propionate-oxidizing bacterium, Pelotomaculum schinkii sp. nov.;

cocultured with Methanospirillum hungatei, and emendeddescription of the genus Pelotomaculum. Int. J. Syst. Evol.Microbiol., 55, 1697-1703 (7 pages).

G al l er t , C .; Winter, J . , ( 2 00 8 ). Propionic a cidaccumulation and degradation during restart of a fu ll-scale anaerobic biowaste digester. Bioresource. Tech.,99 , 170 –178 (9 pages) .

Gerardi, M. H., (2003). Wastewater microbiology series: Themicrobiology of anaerobic digesters. John Wiley and SonsInc., New York.

Ghorbani, F.; Younesi, H.; Ghasempour, S. M.; zinatizadeh, A.A.; Amini. M.; Daneshi, A., (2008). Application of responsesurface methodology for optimization of cadmiumbiosorption in an aqueous solution by Saccharomycescervisiae. Chem. Eng. J., 145, 267–275 (9 pages).

Han, S.K.; Kim, S.H.; Shin, H.S., (2005). UASB treatment ofwastewater with VFA and alcohol generated during hydrogenfermentation of food waste. Proc. Biochem., 40, 2897–2905 (9 pages).

Hooshyari, B.; Azimi, A.; Mehrdadi, N., (2009). Kinetic analysisof enhanced biological phosphorus removal in a hybridintegrated fixed film activated sludge process. Int. J. Environ.Sci. Tech. 6 (1), 149-158 (10 Pages).

Hwang, M. H.; Jang, N. J.; Hyum, S. H.; Kim, I.S., (2004).Anaerobic bio-hydrogen production from ethanolfermentation: The role of pH. J. Biotech., 111, 297–309(13 pages).

Kida, K.; Morimura, S.; Sonoda, Y., (1993). Accumulation ofpropionic acid during anaerobic treatment of distillerywastewater from barley-shochu making. J. Ferment. Bioeng.,75, 213–216 (4 pages).

Kim, I. S.; Hwang, M.H.; Jang, N.J.; Hyun, S.H.; Lee, S.T.,(2004). Effect of low pH on the activity of hydrogen utilizingmethanogen in bio-hydrogen process. Int. J . Hydrogen.Energ., 29, 1133–1140 (8 pages).

Kosaka, T.; Uchiyama, T.; Ishii, S.; Enoki, M.; Imachi, H.;Kamagata, Y.; Ohashi, A.; Harada, H.; Ikenaga, H.; Watanabe,K., (2006). Reconstruction and regulation of the centralcatabolic pathway in the thermophilic propionate-oxidizingsyntroph Pelotomaculum thermopropionicum. J. Bacteriol.,188, 202–210 (9 pages).

Liu, R. R.; Tian, Q.; Yang, B.; Chen, J. H.; Hybrid anaerobicbaffled reactor for treatment of desizing wastewater. Int. J.Environ. Sci. Tech., 7 (1), 111-118 (8 pages).

Mason, R. I.; Gunst, R. F.; Hess, J. L., (2003). Statistical designand analysis of experiments, with applications toengineering and science. 2nd Ed. John Wiley & Sons, NewYork.

Metcalf and Eddy Inc. (2004). Wastewater engineering:treatment and reuse., McGraw–Hill International Editions.4th Ed., New York, Chap. 10, 983-1035 (53 pages).

Montgomery, D.C., (1991). Design and analysis ofexperiments., John Wiley & Sons, New York.

Myers, R.H.; Montgomery, D.C., (2002). Response surfacemethodology., 3rd Ed., John Wiley & Sons, New York.

Nwuche, C. O.; Ugoji E. O., (2008). Effects of heavy metalpollution on the soil microbial activity. Int. J. Environ. Sci.Tech., 5 (3), 409-414 (6 pages).

Nwuche, C. O.; Ugoji E. O., (2010). Effect of co-existingplant specie on soil microbial activity under heavy metalstress. Int. J. Environ. Sci. Tech., 7 (4), 697-704 (8 pages).


95

T. Amani et al.

How to cite this article: (Harvard style)Amani, T.; Nosrati, M.; Mousavi, S. M.; Kermanshahi, R. K., (2011). Study of syntrophic anaerobic digestion of volatile fatty acids usingenriched cultures at mesophilic conditions. Int. J. Environ. Sci. Tech., 8 (1), 83-96.

Plugge, C.; Balk, M.; Stams, A. J. M., (2002). Desulfotomaculumthermobenzoicum subsp. Thermosyntrphicum subsp. nov., athermophilic, syntrophic, propionate-oxidizing, spore-formingbacterium. Int. J. Syst. Evol. Microbiol., 52, 391–399 (9 pages).

Pullammanappallil , P.C.; Chynoweth, D.P.; Lyberatos, G.;Svoronos, S.A., (2001). Stable performance of anaerobicdigestion in the presence of a high concentration ofpropionic acid. Bioresource. Tech., 78, 165–169 (5 pages).

Schink, B.; Stams, A.J.M., (2005). Syntrophism amongprokaryotes. In: Dworkin M. (Ed.), The prokaryotes: anevolving electronic resource for the microbiologicalcommunity, 3rd Ed. Springer, New York.

Schmidt, J. E.; Ahring, B. K., (1995). Interspecies electrontransfer during propionate and butyrate degradation inmesophilic, granular sludge. Appl. Environ. Microbiol., 61(7), 2765–2767 (3 pages).

Shah, B. A.; Shah, A. V.; Singh, R. R. (2009). Sorption isothermsand kinetics of chromium uptake from wastewater usingnatural sorbent material. Int. J. Environ. Sci. Tech., 6 (1),77-90 (14 Pages).

Siegert, I.; Banks, C., (2005). The effect of volatile fatty acidaddition on the anaerobic digestion of cellulose and glucosein batch reactors. Process. Biochem., 40, 3412–3418 (7pages) .

Söteman, S. W.; Ristow, N. E.; Wentzel, M. C.; Ehama, G. A.,(2005). A steady state model for anaerobic digestion of

sewage sludge. Water South Africa, 31 (4), 511–527 (17pages) .

Tatara, M.; Makiuchi, T.; Ueno, Y.; Goto, M.; Sode, K., (2008).Methanogenesis from aceta te and propionate bythermophilic down-flow anaerobic packed-bed reactor.Bioresource Tech., 99, 4786-4795 (10 pages).

Turovskiy, I. S.; Mathai, P. K., (2006). Wastewater sludgeprocessing. New York: Wiley.

Uneo, Y.; Tatara, M., (2008). Microbial population in athermophilic packed-bed reactor for methanogenesis fromvolatile fatty acids. Enzyme Microbiol. Tech., 43, 302–308 (7 pages).

Wang, G.; Mu, Y.; Yu, H.Q. , (2005 ). Response sur faceanalysis to evaluate the influence of pH, temperatureand substra te concentra tion on the a cidogenesis ofsucrose-rich wastewater. Biochem. Eng. J., 23, 175–184(10 pages) .

Wong, B. T.; Show, K. Y.; Lee, D.J.; Lai, J. Y., (2009). Carbonbalance of anaerobic process: Carbon credit. Bioresource.Tech., 100, 1734–1739 (6 pages).

Zinatizadeh, A. A. L.; Mohamed, A. R.; Abdullah, A. Z.;Mashitah, M.D.; Hasnain Isa, M.; Najafpour, G. D., (2006).Process modeling and analysis of palm oil mill effluenttreatment in an up-flow anaerobic sludge fixed filmbioreactor using response surface methodology (RSM). WaterRes., 40, 3193–3208 (16 pages).

AUTHOR (S) BIOSKETCHESAmani, T., Ph.D., scholar, Biotechnology Group, Department of Chemical Engineering, Tarbiat Modares University, P.O. Box 14115-143, Tehran, Iran. Email: [email protected]

Nosrati, M., Ph.D., Assistant Professor, Biotechnology Group,Department of Chemical Engineering, Tarbiat Modares University, P.O.Box 14115-143, Tehran. E-mail: [email protected]

Mousavi, S.M., Ph.D., Assistant Professor, Biotechnology Group,Department of Chemical Engineering, Tarbiat Modares University, P.O.Box 14115-143, Tehran, Iran. E-mail: [email protected]

Kermanshahi, R.K., Ph.D., Full Professor, Department of Biology, Faculty of Basic Sciences, Alzahra University, Tehran, Iran.E-mail: [email protected]

96

study of syntrophic anaerobic digestion of volatile fatty acids using enriched cultures at...

Documents